基于改进混合高斯建模和短时稳定度的遗留物检测算法

被引:5
作者
张超 [1 ]
吴小培 [1 ]
周建英 [1 ]
戚培庆 [1 ]
王营冠 [2 ]
吕钊 [1 ]
机构
[1] 安徽大学计算智能与信号处理教育部重点实验室
[2] 中国科学院上海微系统与信息技术研究所中国科学院无线传感网与通信重点实验室
关键词
遗留物检测; 混合高斯建模; 前景模型; 短时稳定度;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
传统遗留物检测算法存在算法过于复杂和环境适应性差的局限。本文将改进的混合高斯建模方法应用于遗留物检测,利用背景匹配失败时生成的前景模型进行前景匹配并引入短时稳定度指标,在深入挖掘前景模型中包含的遗留物信息和像素点级目标状态信息的基础上对遗留物进行综合判断。文中详细分析了传统方法的性能局限并阐述了新方法中前景模型和短时稳定度的作用原理同时给出了具体的算法流程。多场景下的实验分析表明,增加对前景模型的考察使算法在保留传统方法优点的同时具备了良好的遗留物检测能力,而短时稳定度的引入则能够进一步降低传统方法中前景模型向背景模型转换的风险。对比实验结果中本文方法在表现出良好环境适应性的同时误检团块数明显低于其他方法,算法在复杂背景条件下达到了良好的检测性能。
引用
收藏
页码:1101 / 1111
页数:11
相关论文
共 10 条
[1]  
A distributed surveillance system for detection of abandoned objects in unmanned railway environments. C. Sacchi,C. Regazzoni. IEEE Transactions on Vehicular Technology . 2000
[2]  
Abandoned object detection in highway scene. Huiyuan Fu,Mei Xiang,Huadong Ma,Anlong Ming,Li-ang Liu. Pervasive Computing and Applications (ICPCA),20116th International Conference on . 2011
[3]  
Detecting static objects in busy scenes. Wang J,Ooi W. Technical Report TR99-1730,Department of Com-puter Science,Cornell University . 1999
[4]  
Stationary objectsin multiple object tracking. Guler S,Silverstein J.A,Pushee I.H. Proc.IEEE Conferenceon Advanced Video and Signal Based Surveillance . 2007
[5]  
Left-luggage detection using homographies and simpleheuristics. Auvinet E,Grossmann E,Rougier C,Dahmane M,etc. Proceedings of the 9th IEEEInternational Workshop on Performance Evaluation inTracking and Surveillance (PETS’’06) . 2006
[6]  
Adaptive background mixture models for real-time tracking. Stauffer C,Grimson WEL. Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition . 1999
[7]  
A system for Video Surveillance and Monitoring: VSAM Final Report. R Collins,A Lipton,T.Kanade,et al. Technical Report CMU-RI-TR-00-12 . 2000
[8]  
Abandoned Objects DetectionUsing Double Illumination Invariant Foreground Masks. Li XL,Zhang C,Zhang D. 20thInternational Conference on Pattern Recognition, 2010 ICPR . 2010
[9]  
An Abandoned Objects Management System Based on the Gaussian Mixture Model. Chih-Yang Lin,Wen-Hao Wang. International Conference on Convergence and Hybrid Information Technology . 2008
[10]  
Real-time Video-shot Detection for Scene Surveillance Applications. Stringa E,Regazzoni C S. IEEE Transactions on Image Processing . 2000